From Explainable to Reliable Artificial Intelligence

Autor:	Ivan Vaccari, Vanessa Orani, Sara Narteni, Maurizio Mongelli, Melissa Ferretti, Enrico Cambiaso
Jazyk:	angličtina
Rok vydání:	2021
Předmět:	business.industry Computer science Feature vector Context (language use) Collision Zero (linguistics) Vehicle platooning Logic Learning Machine Reliable AI Skope rules Artificial intelligence Noise (video) business Set (psychology) Logic learning machine
Zdroj:	Lecture Notes in Computer Science ISBN: 9783030840594 CD-MAKE
Popis:	Artificial Intelligence systems are characterized by always less interactions with humans today, leading to autonomous decision-making processes. In this context, erroneous predictions can have severe consequences. As a solution, we design and develop a set of methods derived from eXplainable AI models. The aim is to define “safety regions” in the feature space where false negatives (e.g., in a mobility scenario, prediction of no collision, but collision in reality) tend to zero. We test and compare the proposed algorithms on two different datasets (physical fatigue and vehicle platooning) and achieve quite different conclusions in terms of results that strongly depend on the level of noise in the dataset rather than on the algorithms at hand.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::53367efe36150aaee56d4e07908ac442 http://hdl.handle.net/11583/2966705 Zobrazit plný text záznamu